Kinetics of Interaction of the Myristoylated Alanine-rich C Kinase Substrate, Membranes, and Calmodulin
1997; Elsevier BV; Volume: 272; Issue: 43 Linguagem: Inglês
10.1074/jbc.272.43.27167
ISSN1083-351X
AutoresAnna Arbuzova, Jiyao Wang, Diana Murray, Jaison Jacob, David S. Cafiso, Stuart McLaughlin,
Tópico(s)Protein Structure and Dynamics
ResumoMembrane binding of the myristoylated alanine-rich C kinase substrate (MARCKS) requires both its myristate chain and basic 舠effector舡 region. Previous studies with a peptide corresponding to the effector region, MARCKS-(151–175), showed that the 13 basic residues interact electrostatically with acidic lipids and that the 5 hydrophobic phenylalanine residues penetrate the polar head group region of the bilayer. Here we describe the kinetics of the membrane binding of fluorescent (acrylodan-labeled) peptides measured with a stopped-flow technique. Even though the peptide penetrates the polar head group region, the association of MARCKS-(151–175) with membranes is extremely rapid; association occurs with a diffusion-limited association rate constant. For example,k on = 1011m−1 s−1 for the peptide binding to 100-nm diameter phospholipid vesicles. As expected theoretically,k on is independent of factors that affect the molar partition coefficient, such as the mole fraction of acidic lipid in the vesicle and the salt concentration. The dissociation rate constant (k off) is ∼10 s−1(lifetime = 0.1 s) for vesicles with 107 acidic lipid in 100 mm KCl. Ca2+-calmodulin (Ca2+·CaM) decreases markedly the lifetime of the peptide on vesicles, e.g. from 0.1 to 0.01 s in the presence of 5 ॖm Ca2+·CaM. Our results suggest that Ca2+·CaM collides with the membrane-bound MARCKS-(151–175) peptide and pulls the peptide off rapidly. We discuss the biological implications of this switch mechanism, speculating that an increase in the level of Ca2+-calmodulin could rapidly release phosphatidylinositol 4,5-bisphosphate that previous work has suggested is sequestered in lateral domains formed by MARCKS and MARCKS-(151–175). Membrane binding of the myristoylated alanine-rich C kinase substrate (MARCKS) requires both its myristate chain and basic 舠effector舡 region. Previous studies with a peptide corresponding to the effector region, MARCKS-(151–175), showed that the 13 basic residues interact electrostatically with acidic lipids and that the 5 hydrophobic phenylalanine residues penetrate the polar head group region of the bilayer. Here we describe the kinetics of the membrane binding of fluorescent (acrylodan-labeled) peptides measured with a stopped-flow technique. Even though the peptide penetrates the polar head group region, the association of MARCKS-(151–175) with membranes is extremely rapid; association occurs with a diffusion-limited association rate constant. For example,k on = 1011m−1 s−1 for the peptide binding to 100-nm diameter phospholipid vesicles. As expected theoretically,k on is independent of factors that affect the molar partition coefficient, such as the mole fraction of acidic lipid in the vesicle and the salt concentration. The dissociation rate constant (k off) is ∼10 s−1(lifetime = 0.1 s) for vesicles with 107 acidic lipid in 100 mm KCl. Ca2+-calmodulin (Ca2+·CaM) decreases markedly the lifetime of the peptide on vesicles, e.g. from 0.1 to 0.01 s in the presence of 5 ॖm Ca2+·CaM. Our results suggest that Ca2+·CaM collides with the membrane-bound MARCKS-(151–175) peptide and pulls the peptide off rapidly. We discuss the biological implications of this switch mechanism, speculating that an increase in the level of Ca2+-calmodulin could rapidly release phosphatidylinositol 4,5-bisphosphate that previous work has suggested is sequestered in lateral domains formed by MARCKS and MARCKS-(151–175). The myristoylated alanine-rich C kinase substrate (MARCKS), 1The abbreviations used are: MARCKS, myristoylated alanine-rich protein kinase C substrate; MARCKS-(151–175), peptide corresponding to the basic effector region of bovine MARCKS; K151C-MARCKS-(151–175), peptide with lysine 151 substituted with cysteine that then is labeled with acrylodan; F160C-MARCKS-(151–175), peptide with phenylalanine 160 substituted with cysteine that then is labeled with acrylodan; CaM, calmodulin; Ca2+·CaM, Ca2+-calmodulin complex; PIP2, phosphatidylinositol 4,5-bisphosphate; PC, phosphatidylcholine; PS, phosphatidylserine; PG, phosphatidylglycerol; PE, phosphatidylethanolamine; NBD, 7-nitro-2,1,3-benzoxadiazol-4-yl; MOPS, 4-morpholinepropanesulfonic acid; Fmoc,N-(9-fluorenyl)methoxycarbonyl; C-K7, acrylodan-labeled CKKKKKKK peptide; LUV, large unilamellar vesicle; LUV100, LUV400, and LUV600, large unilamellar vesicles prepared by extrusion through 100-, 400-, and 600-nm filters. a major substrate of protein kinase C in many cell types, is essential for brain development (1Stumpo D.J. Bock C.B. Tuttle J.S. Blackshear P.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 944-948Crossref PubMed Scopus (266) Google Scholar). Although the exact function of MARCKS is not known, this peripheral protein binds to calmodulin (2Graff J.M. Young T.N. Johnson J.D. Blackshear P.J. J. Biol. Chem. 1989; 264: 21818-21823Abstract Full Text PDF PubMed Google Scholar) and actin (3Hartwig J.H. Thelen M. Rosen A. Janmey P.A. Nairn A.C. Aderem A. Nature. 1992; 356: 618-622Crossref PubMed Scopus (620) Google Scholar) and may play a role in secretion, membrane trafficking, and cell motility (reviewed in Refs. 4Aderem A. Cell. 1992; 71: 713-716Abstract Full Text PDF PubMed Scopus (431) Google Scholar and 5Blackshear P.J. J. Biol. Chem. 1993; 268: 1501-1504Abstract Full Text PDF PubMed Google Scholar). In the plasma membranes of macrophages, MARCKS is concentrated in lateral domains (6Rosen A. Keenan K.F. Thelen M. Nairn A.C. Aderem A. J. Exp. Med. 1990; 172: 1211-1215Crossref PubMed Scopus (165) Google Scholar): it colocalizes with actin filaments and protein kinase C-α in nascent phagosomes (7Allen L.A. Aderem A. J. Exp. Med. 1995; 182: 829-840Crossref PubMed Scopus (265) Google Scholar). Binding of MARCKS to either phospholipid vesicles or biological membranes requires both the insertion of its myristate chain into the hydrocarbon interior of the membrane and interaction of its basic effector region with acidic phospholipids (Refs. 8George D.J. Blackshear P.J. J. Biol. Chem. 1992; 267: 24879-24885Abstract Full Text PDF PubMed Google Scholar, 9Taniguchi H. Manenti S. J. Biol. Chem. 1993; 268: 9960-9963Abstract Full Text PDF PubMed Google Scholar, 10Kim J. Shishido T. Jiang X. Aderem A. McLaughlin S. J. Biol. Chem. 1994; 269: 28214-28219Abstract Full Text PDF PubMed Google Scholar, 11Swierczynski S.L. Blackshear P.J. J. Biol. Chem. 1995; 270: 13436-13445Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 12Seykora J.T. Myat M.M. Allen L.A. Ravetch J.V. Aderem A. J. Biol. Chem. 1996; 271: 18797-18802Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 13Swierczynski S.L. Blackshear P.J. J. Biol. Chem. 1996; 271: 23424-23430Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar; reviewed in Refs. 14McLaughlin S. Aderem A. Trends Biochem. Sci. 1995; 20: 272-276Abstract Full Text PDF PubMed Scopus (619) Google Scholar, 15Resh M.D. Cell. Signalling. 1996; 8: 403-412Crossref PubMed Scopus (196) Google Scholar, 16Bhatnagar R.S. Gordon J.I. Trends Cell Biol. 1997; 7: 14-21Abstract Full Text PDF PubMed Scopus (136) Google Scholar). Phosphorylation by protein kinase C (reviewed in Ref.17Newton A.C. J. Biol. Chem. 1995; 270: 28495-28498Abstract Full Text Full Text PDF PubMed Scopus (1472) Google Scholar) introduces three negatively charged phosphates into the effector region, which weakens its electrostatic interaction with acidic lipids and produces desorption of MARCKS from membranes (9Taniguchi H. Manenti S. J. Biol. Chem. 1993; 268: 9960-9963Abstract Full Text PDF PubMed Google Scholar, 10Kim J. Shishido T. Jiang X. Aderem A. McLaughlin S. J. Biol. Chem. 1994; 269: 28214-28219Abstract Full Text PDF PubMed Google Scholar, 11Swierczynski S.L. Blackshear P.J. J. Biol. Chem. 1995; 270: 13436-13445Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 18Thelen M. Rosen A. Nairn A.C. Aderem A. Nature. 1991; 351: 320-322Crossref PubMed Scopus (308) Google Scholar). Ca2+-calmodulin (Ca2+·CaM) binds with high (nanomolar) affinity to the effector region, producing desorption of MARCKS from both phospholipid vesicles and plasma membranes (10Kim J. Shishido T. Jiang X. Aderem A. McLaughlin S. J. Biol. Chem. 1994; 269: 28214-28219Abstract Full Text PDF PubMed Google Scholar, 11Swierczynski S.L. Blackshear P.J. J. Biol. Chem. 1995; 270: 13436-13445Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar). Several experiments suggest that a peptide corresponding to the effector region of bovine MARCKS, MARCKS-(151–175), is a good model for studying the membrane binding of this region. First, membrane binding of both the protein and the peptide depends strongly on the mole fraction of acidic lipid in the membrane and the ionic strength of the solution (10Kim J. Shishido T. Jiang X. Aderem A. McLaughlin S. J. Biol. Chem. 1994; 269: 28214-28219Abstract Full Text PDF PubMed Google Scholar, 19Kim J. Blackshear P.J. Johnson J.D. McLaughlin S. Biophys. J. 1994; 67: 227-237Abstract Full Text PDF PubMed Scopus (146) Google Scholar). Second, CaM binds with high affinity to both the peptide and the effector region of the protein, producing desorption of both molecules from membranes (2Graff J.M. Young T.N. Johnson J.D. Blackshear P.J. J. Biol. Chem. 1989; 264: 21818-21823Abstract Full Text PDF PubMed Google Scholar, 10Kim J. Shishido T. Jiang X. Aderem A. McLaughlin S. J. Biol. Chem. 1994; 269: 28214-28219Abstract Full Text PDF PubMed Google Scholar, 11Swierczynski S.L. Blackshear P.J. J. Biol. Chem. 1995; 270: 13436-13445Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 20McIlroy B.K. Walters J.D. Blackshear P.J. Johnson J.D. J. Biol. Chem. 1991; 266: 4959-4964Abstract Full Text PDF PubMed Google Scholar, 21Verghese G.M. Johnson J.D. Vasulka C. Haupt D.M. Stumpo D.J. Blackshear P.J. J. Biol. Chem. 1994; 269: 9361-9367Abstract Full Text PDF PubMed Google Scholar). Third, protein kinase C phosphorylation inhibits the membrane binding of both MARCKS and the peptide (9Taniguchi H. Manenti S. J. Biol. Chem. 1993; 268: 9960-9963Abstract Full Text PDF PubMed Google Scholar, 10Kim J. Shishido T. Jiang X. Aderem A. McLaughlin S. J. Biol. Chem. 1994; 269: 28214-28219Abstract Full Text PDF PubMed Google Scholar, 18Thelen M. Rosen A. Nairn A.C. Aderem A. Nature. 1991; 351: 320-322Crossref PubMed Scopus (308) Google Scholar, 19Kim J. Blackshear P.J. Johnson J.D. McLaughlin S. Biophys. J. 1994; 67: 227-237Abstract Full Text PDF PubMed Scopus (146) Google Scholar). Fourth, both MARCKS and MARCKS-(151–175) inhibit the phospholipase C-catalyzed hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2). The peptide produces lateral domains enriched in PIP2, but not phospholipase C, which could explain this inhibition (22Glaser M. Wanaski S. Buser C.A. Boguslavsky V. Rashidzada W. Morris A. Rebecchi M. Scarlata S.F. Runnels L.W. Prestwich G.D. Chen J. Aderem A. Ahn J. McLaughlin S. J. Biol. Chem. 1996; 271: 26187-26193Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar). Fig. 1 shows a molecular model of acrylodan-labeled MARCKS-(151–175) bound to a 2:1 phosphatidylcholine/phosphatidylserine (PC/PS) membrane; the model is consistent with the available structural data. Circular dichroism studies of MARCKS-(151–175) suggest that the peptide has little defined secondary structure, either in solution or when bound to a membrane (23Coburn C. Eisenberg J. Eisenberg M. McLaughlin S. Runnels L. Biophys. J. 1993; 64 (abstr.): 60Google Scholar, 24Qin Z. Cafiso D.S. Biochemistry. 1996; 35: 2917-2925Crossref PubMed Scopus (74) Google Scholar). Association of MARCKS-(151–175) with a lipid monolayer held at a constant area increases the surface pressure, which shows that it penetrates the head group region (22Glaser M. Wanaski S. Buser C.A. Boguslavsky V. Rashidzada W. Morris A. Rebecchi M. Scarlata S.F. Runnels L.W. Prestwich G.D. Chen J. Aderem A. Ahn J. McLaughlin S. J. Biol. Chem. 1996; 271: 26187-26193Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar). More detailed EPR studies indicate that membrane-bound MARCKS-(151–175) is in an extended conformation (24Qin Z. Cafiso D.S. Biochemistry. 1996; 35: 2917-2925Crossref PubMed Scopus (74) Google Scholar): the 5 phenylalanine residues penetrate the polar head group region, whereas the first 6 amino acids, which are positively charged, remain outside this region. In Fig. 1, we show acrylodan attached to a cysteine-substituted residue at the N terminus of the peptide; we used this peptide for our kinetic studies. A blue shift in the fluorescence of this polarity-sensitive probe (26Prendergast F.G. Meyer M. Carlson G.L. Iida S. Potter J.D. J. Biol. Chem. 1983; 258: 7541-7544Abstract Full Text PDF PubMed Google Scholar) occurs when the labeled peptide binds to membranes (19Kim J. Blackshear P.J. Johnson J.D. McLaughlin S. Biophys. J. 1994; 67: 227-237Abstract Full Text PDF PubMed Scopus (146) Google Scholar), suggesting that the fluorophore inserts into the membrane as indicated in Fig. 1. This study addresses three questions. First, how fast does MARCKS-(151–175) bind to phospholipid vesicles? Second, what is its lifetime on the vesicle? Third, how rapidly can Ca2+-calmodulin release MARCKS-(151–175) from the membrane? 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylcholine (PC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylserine (PS), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylglycerol (PG), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylethanolamine (PE), N-(7-nitro-2,1,3-benzoxadiazol-4-yl)-egg phosphatidylethanolamine (NBD-PE), 1-acyl-2-[12[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]dodecanoyl]-[phospho-rac-(1-glycerol)] (NBD-PG), and 1-oleoyl-2-[12[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]dodecanoyl]-[phospho-l-serine] (NBD-PS) were purchased from Avanti Polar Lipids (Alabaster, AL). Radiolabeled 1,2-di[1-14C] oleoyl-l-3-phosphatidylcholine ([14C]PC) was purchased from Amersham Corp., and 6-acryloyl-2-dimethylaminonaphthalene (acrylodan) from Molecular Probes, Inc. (Eugene, OR). K151C-MARCKS-(151–175) has the sequence acrylodan-CKKKKRFSFKKSFLKSGFSFKKNKK-amide and was obtained from Boehringer Mannheim (Mannheim, Germany; purity > 967). The standard buffer contained 100 mm KCl and 1 mmMOPS, pH 7; buffers with 200 mm KCl were used in some experiments. Bovine brain calmodulin (high purity) was purchased from Calbiochem, and hog brain calmodulin was obtained from Boehringer Mannheim. F160C-MARCKS-(151–175), a peptide corresponding to the basic effector region of bovine MARCKS with the ninth phenylalanine changed to cysteine (acetyl-KKKKKRFSCKKSFKLSGFSFKKNKK-amide) and this cysteine labeled with acrylodan, was synthesized on a Gilson Multiple Peptide system (AMS 422) using a Rink amide p-methylbenzhydrylamine resin and Fmoc chemistry. N-terminal acetylation of the resin-bound peptide was carried out using acetic anhydride and diisopropylethylamine. The peptide was cleaved by reacting the resin with trifluoroacetic acid/ethanedithiol/thioanisole/anisole/water (89:3:3:2:3) for 6 h. The resulting peptide solution was filtered, precipitated in cold ether, dried, and reacted with 1.2 eq of acrylodan in dimethylformamide for 12 h; the reaction mixture was then quenched with water and lyophilized. The lyophilized mixture was dissolved in water, applied to a reverse-phase C18 high pressure liquid chromatography column, and eluted with a 10–457 gradient of acetonitrile in water using a flow rate of 3 ml/min; the pure labeled peptide eluted in 26–27 min. The purified peptides were examined using electrospray/ionization mass spectrometry and yielded m/zvalues of 3304, consistent with the expected molecular mass for the derivatized peptide. C-K7 (acrylodan-CKKKKKKK-amide) was synthesized on an Applied Biosystems Model 431A peptide synthesizer using Fmoc chemistry at the Center for Analysis and Synthesis of Macromolecules, State University of New York (Stony Brook, NY). Cleavage of C-K7 from the resin, purification, and labeling with acrylodan were performed as described above for F160C-MARCKS-(151–175). The initial concentration of lipids in chloroform was measured on a Cahn electrobalance, a method that gives the same results as phosphate analysis (27Kim J. Mosior M. Chung L.A. Wu H. McLaughlin S. Biophys. J. 1991; 60: 135-148Abstract Full Text PDF PubMed Scopus (234) Google Scholar). Trace amounts of [14C]PC were added to the lipid mixture to determine the final lipid concentration. Lipid mixtures were dried on a rotary evaporator and resuspended in the standard buffer to obtain multilamellar vesicles. Large unilamellar vesicles (LUV) were made by extruding multilamellar vesicles 10 times through a stack of two polycarbonate filters (100-, 400-, or 600-nm pore diameter) after five freeze-thaw cycles (28Hope M.J. Bally M.B. Webb G. Cullis P.R. Biochim. Biophys. Acta. 1985; 812: 55-65Crossref PubMed Scopus (2020) Google Scholar). LUV100 (produced using 100-nm pore filters) are unilamellar and have a well defined size distribution, with an average diameter of 100 ± 5 nm (28Hope M.J. Bally M.B. Webb G. Cullis P.R. Biochim. Biophys. Acta. 1985; 812: 55-65Crossref PubMed Scopus (2020) Google Scholar, 29Mui B.L. Cullis P.R. Evans E.A. Madden T.D. Biophys. J. 1993; 64: 443-453Abstract Full Text PDF PubMed Scopus (188) Google Scholar, 30Koelchens S. Ramaswami V. Birgenheier J. Nett L. O'Brien D.F. Chem. Phys. Lipids. 1993; 65: 1-10Crossref PubMed Scopus (63) Google Scholar). LUV400 and LUV600 preparations include both unilamellar and multilamellar vesicles with broader size distributions: LUV400 produced at 50–100 p.s.i. extrusion pressure have average diameters of 200 ± 20 nm; LUV600 have a complex bimodal size distribution, with diameters of ∼100–200 and 400 nm (30Koelchens S. Ramaswami V. Birgenheier J. Nett L. O'Brien D.F. Chem. Phys. Lipids. 1993; 65: 1-10Crossref PubMed Scopus (63) Google Scholar, 31Ertel A. Marangoni A.G. Marsh J. Hallett F.R. Wood J.M. Biophys. J. 1993; 64: 426-434Abstract Full Text PDF PubMed Scopus (75) Google Scholar). Unless specified, preparations and measurements were done at room temperature (22–23 °C). We consider the association of the peptide (P) with the vesicle (V) to form PV. In the simplest case, partitioning occurs in a single step with a second-order association rate constant (k on) and a first-order dissociation rate constant (k off) (Equation 1), P+V⇌koffkonPV;d[PV]/dt=kon[V][P]−koff[PV]Equation 1 where [P] is the concentration of free peptide in the aqueous phase. When the concentration of lipid is much higher than the concentration of peptide, partitioning of the peptide onto the vesicles does not affect the overall affinity of the vesicles for the peptide, [V] is a constant, and association is a pseudo first-order process. Under these conditions, the solution to Equation 1 is as follows (Equation 2), [PV]=[PV]∞×(1−e−t/τ)Equation 2 where [PV]∞ is the concentration of peptide partitioned onto the vesicles at equilibrium and τ is the relaxation time of the process. The reciprocal of the relaxation time (1/τ) is as follows (Equation 3). 1/τ=kon[V]+koffEquation 3 We plot experimental data as 1/τ versus either [V] or [L], where [L] is the concentration of accessible lipid,i.e. the lipid in the outer leaflet of the vesicles. We assume that the accessible lipid concentration is half the total lipid concentration. The association rate constant (k on) can be obtained from the slope of 1/τversus [L], which equals k on/ν, where ν is the number of lipid molecules composing the outer leaflet of the vesicle. ν = 4πR V2/A L, where R V is the radius of a vesicle andA L = 0.7 nm2 is the area/lipid molecule (32McIntosh T.J. Simon S.A. Biochemistry. 1986; 25: 4058-4066Crossref PubMed Scopus (265) Google Scholar). The diffusion-limited association rate constant (k diff) is the highest possible value fork on. For a peptide partitioning onto a vesicle,k diff = 4πN A (R V +R P)(D V +D P), where N A is Avogadro's number, R V +R P is the sum of the radii of the vesicle and the peptide, and D V + D Pis the sum of diffusion coefficients of the vesicle and the peptide (33Noyes R.M. Prog. React. Kinet. 1961; 1: 129-160Google Scholar, 34Berg H.C. Random Walks in Biology. Princeton University Press, Princeton, NJ1983: 17-27Google Scholar, 35Schwarz G. Biophys. Chem. 1987; 26: 163-169Crossref PubMed Scopus (15) Google Scholar). The vesicle is much larger than the peptide (R V + R P ≈R V), and the diffusion coefficient of the vesicle is much lower than that of the peptide (D V + D P ≈D P); hence, Equation 4 follows. kdiff=4πNARVDPEquation 4 Kinetic measurements were performed on an SLM-AMINCO spectrofluorometer with a stopped-flow attachment (Milliflow SLM-AMINCO, Rochester, NY). The dead time was 2–4 ms as determined by measuring the reaction of pyranine (Molecular Probes, Inc.) with bovine carbonic anhydrase (Sigma) as described by the manufacturer. Acrylodan fluorescence was monitored at 455–460 nm with excitation at 362–370 nm and slit widths of 4–8 nm. All solutions were degassed. In a typical experiment, 100 nm peptide in buffer was mixed rapidly with the vesicle solution in a stopped-flow chamber (volume = 32 ॖl). An average of at least 10 shots was fit with a single exponent (least squares method). The quality of the fit was checked by examining the residuals between the data points and the fit. We determined the dissociation rate constant (k off) in three ways: as the y axis intercept of a plot of 1/τversus [V] (Equation 3), from the kinetics of peptide transfer from donor to acceptor vesicles (see below), and as the ratio of the association rate constant and the molar partition coefficient (see Equation 7). In the second approach, we mixed acceptor vesicles (V*) with a solution containing peptide adsorbed to donor vesicles (PV). We describe the redistribution of the peptide from donor to acceptor vesicles using Equation 5, PV+V*⇌konkoffP+V+V*⇌k*offk*onPV*+VEquation 5 where k*on andk*off are the association and dissociation rate constants for the peptide binding to acceptor vesicles, respectively. As discussed elsewhere (36Kuzelova K. Brault D. Biochemistry. 1994; 33: 9447-9459Crossref PubMed Scopus (55) Google Scholar), the relaxation time of the transfer process (τtr) is dominated by k offif three conditions are met: k*on[V*] ≫ k on[V], k*off≪ k off, andk*on[V*] ≫ k off. The first condition is met because [V*] ≫ [V] andk*on ≈ k on; the second condition is met because the acceptor vesicles have a higher mole percent acidic lipid, which increases the affinity of the peptide for V* compared with V; and the third condition is met because the peptide binds strongly to the vesicles (k off < 100 s−1) and association is rapid (we used an excess of the acceptor vesicles, k*on[V*] > 1000 s−1). Under these conditions, Equation 6 follows. koff=1/τtrEquation 6 We measured the transfer of the peptide from donor to acceptor vesicles by resonance energy transfer assay. The acrylodan label on the peptide (excitation at 370 nm) fluoresces at 520 nm in buffer and at 460 nm when bound to membranes; the NBD label on the lipid (excitation at 460 nm) emits at 540 nm. Thus, acrylodan-labeled MARCKS-(151–175) produces little fluorescence at 460 nm if it is either free in solution or bound to a membrane containing NBD-labeled lipid. In a typical experiment, one syringe contained peptide premixed with donor vesicles under conditions where most of the peptide was bound; the other syringe contained an excess of acceptor vesicles containing a high mole percent acidic lipid (usually 1:1 PC/PS; molar partition coefficient of peptide onto these vesicles of K > 107m−1 (19Kim J. Blackshear P.J. Johnson J.D. McLaughlin S. Biophys. J. 1994; 67: 227-237Abstract Full Text PDF PubMed Scopus (146) Google Scholar)). Either donor or acceptor vesicles contained 1 mol 7 NBD-labeled PE, PS, or PG to quench acrylodan fluorescence. We obtained τtr from a single exponential fit of the acrylodan fluorescence as a function of time. The relaxation time was independent of vesicle concentration when the ratio of acceptor to donor vesicles was >8, indicating that the three conditions are met and peptide transfer occurs through the aqueous phase. We obtained our third independent estimate of the dissociation rate constant (k off) using the measured values ofk on/ν and the molar partition coefficient (K) (Equation 7). koff=kon/(νK)Equation 7 The molar partition coefficient is the reciprocal of [L] that binds 507 of the peptide. The determination of K is described in 舠Appendix 舡 and elsewhere (19Kim J. Blackshear P.J. Johnson J.D. McLaughlin S. Biophys. J. 1994; 67: 227-237Abstract Full Text PDF PubMed Scopus (146) Google Scholar, 37Peitzsch R.M. McLaughlin S. Biochemistry. 1993; 32: 10436-10443Crossref PubMed Scopus (462) Google Scholar). We also used resonance energy transfer to measure the transfer of acrylodan-labeled MARCKS-(151–175) from vesicles to Ca2+·CaM. One syringe contained peptide bound to vesicles containing 1 mol 7 NBD-labeled PS or PG to quench the acrylodan fluorescence of the peptide. The other syringe contained Ca2+·CaM in 100 mm KCl and 1 mmMOPS, pH 7, and 50 or 100 ॖm CaCl2. We mixed the contents of the two syringes and monitored the increase in acrylodan fluorescence that occurred when peptides moved from vesicles to calmodulin (excitation at 370 nm and emission at 460 nm). We designed the experiments so that initially, most of the peptide was bound to vesicles, and after mixing with Ca2+·CaM, most of the peptide was bound to Ca2+-calmodulin. Control experiments showed that the relaxation time for MARCKS-(151–175) transfer from the membrane to CaM (τCaM) was independent of both lipid concentration (for example, 50–200 ॖm 5:1 PC/PS vesicles) and Ca2+ concentration (20–200 ॖm added CaCl2). Adding the same concentration of CaCl2 without calmodulin to the peptide/vesicle mixture did not change the acrylodan fluorescence. The kinetics of peptide transfer to Ca2+·CaM did not depend strongly on the peptide concentration in the range of 20–200 nm. Most experiments were done using bovine calmodulin from Calbiochem. We repeated several experiments using porcine CaM from Boehringer Mannheim and obtained qualitatively similar results. We measured the kinetics of peptide partitioning onto vesicles by rapidly mixing the acrylodan-labeled peptide K151C- or F160C-MARCKS-(151–175) with phospholipid vesicles in a stopped-flow chamber. Fig.2 A shows measurements on samples containing 100 nm peptide and four different concentrations of 5:1 PC/PS vesicles. The data illustrate three points. First, the acrylodan fluorescence increases when the peptide associates with vesicles. The steady-state fluorescence values correlate with the fraction of F160C-MARCKS-(151–175) bound to membrane: in the experiment illustrated by curve a, ∼307 of the peptide was bound to vesicles ([L] = 1.2 ॖm), whereas incurve d, ∼907 of F160C-MARCKS-(151–175) was bound to vesicles ([L] = 25 ॖm). Second, the increase in fluorescence over time can be described well by a single exponential function (Equation 2); the solid curves are the best single exponential fits. Third, the relaxation time (τ), determined from the single exponential fit, decreases as the lipid concentration increases. As discussed in 舠Appendix ,舡 the relaxation time was independent of the peptide concentration in all the experiments presented in Figs.2, 3, 4, 5.Figure 3Value of k on depends on the radius of the vesicle. The values of 1/τ were obtained as described in the legend to Fig. 2. The concentration of vesicles ([V]) = [L]/ν; the values of ν are given in column A of TableI. The circles show the values of 1/τ for K151C-MARCKS-(151–175) partitioning onto 5:1 PC/PS 100-nm vesicles (LUV100) in 100 mm KCl and 1 mmMOPS, pH 7. The solid line has the slopek on = 1 × 1011m−1 s−1. The dashed line, taken from Fig. 2 B, represents the linear fit for LUV600 with k on = 3 × 1011m−1 s−1 (TableI).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4A, kinetics for binding of C-K7 to 1:1 PC/PS LUV600. 100 nmacrylodan-labeled C-K7 was mixed rapidly with the vesicles in a stopped-flow chamber (buffer: 100 mm KCl and 1 mm MOPS, pH 7). The concentration of the accessible lipid was 5 ॖm (curve a, ○), 10 ॖm(curve b, □), 15 ॖm (curve c, ▵), or 50 ॖm (curve d, ▿). The acrylodan fluorescence was monitored at 460 nm (excitation at 370 nm); each experimental curve is an average of at least 10 shots. The solid curves show single exponential fits with the relaxation time τ = 190 ms (curve a), τ = 170 ms (curve b), τ = 120 ms (curve c), or τ = 40 ms (curve d). a.u., arbitrary units. B, 1/τ increases linearly as the concentration of accessible lipid increases. Thesymbols show the values of 1/τ obtained experimentally for C-K7 association with 1:1 PC/PS (•) or 2:1 PC/PS (▿) LUV600 in 100 mm KCl and 1 mm MOPS, pH 7. The solid line shows the best linear fit withk on/ν = 1.1 × 106m−1 s−1.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 5Estimation of the dissociation rate constant (k off) from the y axis intercept of a plot of 1/τ versus [L]. The kinetics of K151C-MARCKS-(151–175) association with 10:1 PC/PS LUV100were measured as described in the legend to Fig. 2. Thecircles show the values of 1/τ obtained experimentally. The solid line shows the best linear fit of the data with the slope k on/ν = 2 × 106m−1 s−1 and they axis intercept k off = 8 s−1.View Large Image
Referência(s)